Methodology for illness detection by data analysis techniques

The research aims to develop information technology for identifying problematic health conditions by analyzing measurement data. The literature review highlights various approaches to medical diagnostics, including statistical and machine-learning models that predict the risk of adverse outcomes based on patient data. Developed information technology focuses on data classification and sufficiency, ensuring objective and relevant data is collected. The technology involves expert-defined rules for analysis, aiding in generating patient diagnosis candidates. The proposed information system comprises four components: data source, data storage, diagnosis module, and data sink. A comprehensive data storage structure is designed to store and manage data related to diagnoses and parameters efficiently. The rule set generation block prototype includes obtaining parameters and transforming algorithms into programming functions. A case study focuses on a diagnostic tool for assessing PTSD using an internationally recognized questionnaire. Telegram bot is selected as the data source due to its anonymity, flexibility, and automated data collection capabilities. The database structure is designed to accommodate questionnaire modifications and continue data collection. The implemented analytical system effectively categorizes individuals' states based on their responses. Overall, the research demonstrates the potential of information technology and the proposed information system to provide effective and user-friendly health diagnostics, aiding in timely medical interventions and improving population well-being.